Author Topic: EFI and Ethanol  (Read 3417 times)

0 Members and 1 Guest are viewing this topic.

NH Oldguy

  • Scooter
  • **
  • Posts: 18
  • Karma: 0
on: May 26, 2013, 08:12:39 pm
This is not a rant about ethanol, or E-10 gasoline.  It's a fine hydrocarbon, and I'm quite fond of it. 

However, after reading Bradey's thread on EFI cleaning and overhaul, and mulling my earlier thread about EFI stumble off idle, I had a thought about how well the Enfield ECU may or may not cope with fuel variation.  I'm a mechanical engineer and ignorant of electronics, so I hope some of you may be better informed on the Bullet's fuel management system and its functional capabilities.

Here's my question.  Since ethanol is somewhat less energetic than gasoline on a volumetric (per gallon) basis, a gallon of E-10 has less chemical energy in its hydrocarbon bonds than does a gallon of straight gas.  E-10 therefore requires less air to combust completely than does a similar volume of straight gas.  Said the other way 'round, a given flow of air will burn a greater volume of E-10 than of gasoline.

The various sensor inputs allow the ECU to compute the flow rate (both volume and mass) of air going into the engine at any set of operating conditions, and the ECU's "map" then tells the fuel pump how much volume of fuel to deliver to the injector nozzles.  But the engine needs a HIGHER flow of E-10 for any given air flow than it does of straight gas.  If the engine gets the SAME flow of E-10 as it does of gas, then there will be excess oxygen carried out the exhaust - ie: the engine is running lean.

I know there is a Lambda (O2) sensor in the exhaust head pipe.  Is this just to confirm that SOME O2 carries through (ie; all the fuel was burned) for emissions control purposes?  Or does it also tell the ECU to increase the fuel pump delivery to richen the mixture? 

I have my doubts about the latter.  Is the ECU algorithm sophisticated enough to respond to fuel mixture variations?  I can't see how it knows what is in the tank, other than via the Lambda sensor.  Your thoughts?



ace.cafe

  • Grand Gearhead
  • *****
  • Posts: 14,457
  • Karma: 1
  • World leaders in performance/racing Bullets
Reply #1 on: May 26, 2013, 09:12:42 pm
There is a very large difference in air/fuel ratio requirement with E85, but it should cope fine with E10.
Home of the Fireball 535 !


Ekatus Atimoss

  • Scooter
  • **
  • Posts: 39
  • Karma: 0
Reply #2 on: May 26, 2013, 09:30:50 pm
I'm not an engineer, not even a mechanical one. As we had a huge E10 discussion over here where I live I feel compelled to reply. It would be a surprise to me if the RE engineers had E10 / ethanol on their minds. Despite of the ethical discussion (competition of E10 with food), ethanol is aggressive to some sorts of plastic, and while car makers (in europe) have anticipated the increase of ethanol mix in fuel early enough to prepare their cars, I doubt that the same is true for a bike like the RE. I've been told that ethanol is big in the U.S., so it would be interesting to know how all other bikes cope with it. According to Wikipedia, the head of engine R&D at BMW considers E10 to be harmful to the engine, increasing the amount of water during combustion, thinning the oil and shorten the oil change interval.
 
The EFI O2 sensor is from what I've learned a single point, not covering all operations of the engine. The Autotune kit for the PowerCommander V comes with a multi-point sensor. So to me the sole purpose of the O2 sensor is to make the EFI pass some exhaust emission checks. Pretty much similar nonsense like noise emission checks.
5% ethanol is standard for regular fuel in Germany, E10, as the name implies, got 10%. Most high octane premium fuels don't have any ethanol mix at all. That's what I prefer for the EFI to be on the safe side. Even if the engine characteristic and compression might not justify that ;-)

kind regards
Michael
dopo notte, atra e funesta, splende in ciel più vago il sole e di gioia empie la terra
2012 C5 built for comfort, not for speed (click for pictures)


ERC

  • Grand Gearhead
  • *****
  • Posts: 1,810
  • Karma: 0
Reply #3 on: May 26, 2013, 10:08:28 pm
Here in U.S. most all fuel has 10% in it. Doesn't matter what the octane rating is.  ERC
2-57 Apaches, 2-57 Trailblazers, 60 Chief, 65 Interceptor, 2004 Bullet, 612 Bullet chopped.


barenekd

  • Grand Gearhead
  • *****
  • Posts: 5,516
  • Karma: 0
Reply #4 on: May 26, 2013, 10:29:44 pm
My bike is barely on the side of rich on a dyno and I ran the crap gas in it for it's whole life. No problems. I don't know if the EFI measures mixtures or just air pressure, but it certainly took care of altitude changes quite nice. Mine would occasionally cut out coming off idle, but it wasn't very often.
Bare
2013 Moto Guzzi V7 Racer
2011 Black Classic G5 (RIP)
I refuse to tiptoe through life only to arrive safely at death
http://www.controllineplans.com


Arizoni

  • Grand Gearhead
  • *****
  • Posts: 7,412
  • Karma: 2
  • "But it's a dry heat here in Arizona
Reply #5 on: May 27, 2013, 01:06:08 am
For what it's worth, the ECU doesn't vary the fuel pump pressure to control the fuel flow.  The pump runs at a constant 294 kPa (42.6 psi) pressure.
The ECU controls  the electrical pulse to the fuel injector adjusting how often/long it remains open.

As for the 10 percent alcohol fuel, that's all we've had for several years here in Phoenix, Az and it hasn't caused a problem yet.

I'm just guessing but I know the Fuel Injection system and ECU were designed (and built?) in Japan.  They've been dealing with this gasoline/alcohol fuel mix for many years now so I'm betting they set up the maps to cope with this issue, at least for the 10 percent mix.
Jim
2011 G5 Deluxe
1999 Miata 10th Anniversary


MrMike

  • Bulleteer
  • ***
  • Posts: 131
  • Karma: 0
  • ASA Lives!
Reply #6 on: May 27, 2013, 01:29:51 am
Good stuff to know.  I was contemplating running ethinol free fuel in mine.  Souns like it would not help???
2011 G5 Deluxe w/Kozi Rocket
2009 V-Star 950 Tourer w/Velorex 562
1999 Honda VT 1100 T w/Dauntless Tandem


gashousegorilla

  • Grand Gearhead
  • *****
  • Posts: 4,485
  • Karma: 0
Reply #7 on: May 27, 2013, 04:22:19 am


I know there is a Lambda (O2) sensor in the exhaust head pipe.  Is this just to confirm that SOME O2 carries through (ie; all the fuel was burned) for emissions control purposes?  Or does it also tell the ECU to increase the fuel pump delivery to richen the mixture? 

I have my doubts about the latter.  Is the ECU algorithm sophisticated enough to respond to fuel mixture variations?  I can't see how it knows what is in the tank, other than via the Lambda sensor.  Your thoughts?

  Yes and no. The ECU is sophisticated enough to adjust AFR's based on input from ALL the sensors, as long as it's within the look up tables programmed in the MAP.  The o2 sensor is a heated.... takes around a minute or so to start giving readings to the ECU.......Narrow band sensor.  It sends voltage readings back to the UCE....The narrow band sensors are more like a switch, where they reach a certain voltage and "switch" to the rich or lean side of Stoichiometric. A wide band o2 sensor is more precise..... voltage readings equate to AFR's. The ECU or after market devise see's these readings and adjusts. It's all electrical and mostly involves voltage readings, and all the sensors and has no idea whats in your tank. The MAP on a stock ECU is supposedly very broad for a stock bike. From sea level to 13000 ft. And i'm sure they likely programmed that MAP  using a fuel they expected to see in the Market the MAP was created for.


 
For what it's worth, the ECU doesn't vary the fuel pump pressure to control the fuel flow.  The pump runs at a constant 294 kPa (42.6 psi) pressure.
The ECU controls  the electrical pulse to the fuel injector adjusting how often/long it remains open.

As for the 10 percent alcohol fuel, that's all we've had for several years here in Phoenix, Az and it hasn't caused a problem yet.

I'm just guessing but I know the Fuel Injection system and ECU were designed (and built?) in Japan.  They've been dealing with this gasoline/alcohol fuel mix for many years now so I'm betting they set up the maps to cope with this issue, at least for the 10 percent mix.

   
  Yup, probably the best too. Keihin sensors and ECU..... same as in all of those very  reliable  Accords and Civics.
« Last Edit: May 27, 2013, 05:05:40 am by gashousegorilla »
An thaibhsí atá rattling ag an doras agus tá sé an diabhal sa chathaoir.


gremlin

  • Grand Gearhead
  • *****
  • Posts: 1,873
  • Karma: 0
  • "Do one thing each day that scares you"
Reply #8 on: May 27, 2013, 06:45:51 am
..........ethanol......  It's a fine hydrocarbon, and I'm quite fond of it. 

Me Too !

+1 to GHG, the EFI hardware is way more capable than we know.  As for the narrowband sensor --> standard practise is to widen the fuel pulse until the O2 signal disappears for 3 consecutive hits, then begin narrowing it until the O2 signal just returns. 

wash, rinse, repeat .....  thousands of times per minute.

« Last Edit: May 27, 2013, 06:58:12 am by gremlin »
1996 Trophy 1200
2009 Hyosung GV250
2011 RE B5


NH Oldguy

  • Scooter
  • **
  • Posts: 18
  • Karma: 0
Reply #9 on: May 27, 2013, 05:30:13 pm
Thanks to all for your helpful input, especially GHG and Gremlin.  I now have 5 areas for experimentation, not necessarily in sequence, after laying down a repeatable baseline test run for comparison:

1.  Try a dose of Techron in a tank of E-10,
2.  Try a tank of straight gas (located a source nearby),
3.  Advance the TPS idle output voltage,
4.  Tweak with the idle air bypass screw, and
5.  Try temporarily disabling the lambda sensor.

Are there potentially harmful downsides to #5?  If, as Ekatus thinks, the ECU control strategy on lambda is simply to control emissions, I would think I could test this case for impact on my problem (engine stumble upon throttling up from idle) without harm.  If, as Gremlin says, the ECU is using lambda feedback to keep nudging the AFR just barely lean of stoiciometric, then disabling the sensor might cause the ECU to drive the fuel pump output to the limit of the map range, one way or the other (max or min) - much smoke or knocka-knock?

I appreciate your further input on this, and I will post my experimental findings as I make them - now that last night's snow is gone!   


gremlin

  • Grand Gearhead
  • *****
  • Posts: 1,873
  • Karma: 0
  • "Do one thing each day that scares you"
Reply #10 on: May 27, 2013, 06:23:41 pm
As much as I enjoy lite banter .....  I hafta ask this:

Are you sure your idle is high enough ?
1996 Trophy 1200
2009 Hyosung GV250
2011 RE B5


NH Oldguy

  • Scooter
  • **
  • Posts: 18
  • Karma: 0
Reply #11 on: May 27, 2013, 09:10:37 pm
The correct first question, thank you gremlin. 

On infrequent occasion the engine will die when idling at a long light after fully warming up, so my idle speed could use a bit of a boost.  This happens when the temperature-dependent output of the lambda sensor does a feedback oscillation with the ECU, running the idle speed slowly (5-10 seconds or so) up and down over probably a couple hundred rpm.

So I will raise the idle as experiment #0, after baselining the present stumble behavior more firmly.  Thanks again.


gashousegorilla

  • Grand Gearhead
  • *****
  • Posts: 4,485
  • Karma: 0
Reply #12 on: May 28, 2013, 02:52:23 am


 

wash, rinse, repeat .....  thousands of times per minute.



   Love that.....


Thanks to all for your helpful input, especially GHG and Gremlin.  I now have 5 areas for experimentation, not necessarily in sequence, after laying down a repeatable baseline test run for comparison:

1.  Try a dose of Techron in a tank of E-10,
2.  Try a tank of straight gas (located a source nearby),
3.  Advance the TPS idle output voltage,
4.  Tweak with the idle air bypass screw, and
5.  Try temporarily disabling the lambda sensor.

Are there potentially harmful downsides to #5?  If, as Ekatus thinks, the ECU control strategy on lambda is simply to control emissions, I would think I could test this case for impact on my problem (engine stumble upon throttling up from idle) without harm.  If, as Gremlin says, the ECU is using lambda feedback to keep nudging the AFR just barely lean of stoiciometric, then disabling the sensor might cause the ECU to drive the fuel pump output to the limit of the map range, one way or the other (max or min) - much smoke or knocka-knock?

I appreciate your further input on this, and I will post my experimental findings as I make them - now that last night's snow is gone!   


   #3... I would do in baby steps... Live, with the bike running, so you can hear and smell for any changes. Then test ride and take some plug readings.

   #4  Yes.......

    #5?   Will likely just just throw and store a code. The ECU will probably just default to a safe rich setting... Enough to get you home.... and your bike will run like crap.  Taking it out and cleaning it may be a much better idea.  As mentioned earlier.... The ECU does not control fuel pump pressure, it only brings it on and off. Pressure is constant and circulates back to the tank.  the ECU controls injector duty cycle...... Length of time and amount.
« Last Edit: May 28, 2013, 02:57:51 am by gashousegorilla »
An thaibhsí atá rattling ag an doras agus tá sé an diabhal sa chathaoir.


Gypsyjon

  • Grease Monkey
  • ****
  • Posts: 454
  • Karma: 0
  • 201o G5 Deluxe Black & Chrome
Reply #13 on: May 28, 2013, 03:06:51 am
My G5 will die at lights if I do not keep an eye on it. RPM drops down to very low 500s. I just blip the throttle once in a while like I did on my old panheads and everything is fine. In fact I can often get the blip just right so that it puts out just one THUMP...which is pretty cool!


AwL

  • Scooter
  • **
  • Posts: 58
  • Karma: 0
Reply #14 on: May 28, 2013, 04:54:06 am
Here is a link to a site that lists gas stations around the U.S. and Canada that have ethanol-free gas.  http://pure-gas.org/